This AI model provider switchboard is private. Enter your authorization PIN to unlock the dashboard.
Centralized LLM Model Provider & Switchboard
inst-loading...
Try searching for a different keyword or reset active filters.
Fast reference for connecting custom scripts, apps, and IDEs to your gateway
This gateway exposes standard OpenAI-compatible v1 endpoints. You can use it as a drop-in replacement for OpenAI in any SDK or library.
https://llm.freeutils.in/v1/chat/completions
Chat completions proxy (streaming & non-streaming)
https://llm.freeutils.in/v1/models
Lists all currently enabled models in your catalog
https://llm.freeutils.in/api/instance/reset
Forces fresh instance identity to bypass OpenCode limits
apiKey (e.g. "sk-gateway"). The gateway handles authorization automatically.
Send a JSON payload to POST /v1/chat/completions. Below are the key parameters and how to switch models:
| Field | Type | Required | Description |
|---|---|---|---|
model |
string | Required |
The target Model ID. Change this to switch models! Examples: mimo-v2.5-free, deepseek-v4-flash-free, nemotron-3-ultra-free, claude-3-5-sonnet.
|
messages |
array | Required | Array of conversation turns: [{"role": "system"|"user"|"assistant", "content": "..."}]. |
stream |
boolean | Optional | Set to true for Server-Sent Events (SSE) token streaming. Default: false. |
temperature |
number | Optional | Sampling randomness (0.0 to 1.0). Use 0.2 for code, 0.7 for chat. Default: 0.3. |
max_tokens |
integer | Optional | Maximum response length token cap (e.g. 4096, 8192). Default: 4096. |
tools |
array | Optional | OpenAI-standard tool & function definitions for agentic workflows. |
from openai import OpenAI
# Initialize client pointing to your personal gateway
client = OpenAI(
base_url="https://llm.freeutils.in/v1",
api_key="sk-freeutils-personal" # Any string works
)
# 1. Non-Streaming Request
response = client.chat.completions.create(
model="mimo-v2.5-free", # Change to any enabled model!
messages=[
{"role": "system", "content": "You are an expert Python engineer."},
{"role": "user", "content": "Write a clean LRU cache in Python."}
],
temperature=0.2,
max_tokens=2048,
stream=False
)
print(response.choices[0].message.content)
# 2. Streaming Request (Real-time token output)
stream_response = client.chat.completions.create(
model="deepseek-v4-flash-free",
messages=[{"role": "user", "content": "Explain async/await in 3 bullet points."}],
stream=True
)
for chunk in stream_response:
content = chunk.choices[0].delta.content or ""
print(content, end="", flush=True)import OpenAI from 'openai';
const openai = new OpenAI({
baseURL: 'https://llm.freeutils.in/v1',
apiKey: 'sk-freeutils-personal',
});
async function main() {
const stream = await openai.chat.completions.create({
model: 'mimo-v2.5-free', // Change model ID as needed
messages: [{ role: 'user', content: 'Generate a TypeScript generic debounce function.' }],
stream: true,
});
for await (const chunk of stream) {
process.stdout.write(chunk.choices[0]?.delta?.content || '');
}
}
main();curl -X POST https://llm.freeutils.in/v1/chat/completions \
-H "Content-Type: application/json" \
-d '{
"model": "mimo-v2.5-free",
"messages": [
{ "role": "user", "content": "Explain binary search in simple words." }
],
"temperature": 0.3,
"stream": false
}'You can connect your gateway to popular AI coding tools by setting the custom OpenAI Base URL:
Provider: OpenAI Compatible
Base URL: https://llm.freeutils.in/v1
API Key: sk-freeutils
Model ID: mimo-v2.5-free
{
"models": [
{
"title": "Gateway MiMo Free",
"provider": "openai",
"model": "mimo-v2.5-free",
"apiBase": "https://llm.freeutils.in/v1",
"apiKey": "sk-freeutils"
}
]
}Test completions directly through the provider gateway
inst-...
// Response will stream here...